litellm - ✅(Solved) Fix [Bug]: Sessions are no longer correctly grouped together in the Spend Logs UI [1 pull requests, 1 comments, 2 participants]

litellm2026-03-09 14:46:22

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

BerriAI/litellm#23157•Fetched 2026-04-08 00:38:15

View on GitHub

Comments

Participants

Timeline

Reactions

Author

PClmnt

Participants

nehaaprasad

PClmnt

Timeline (top)

labeled ×3mentioned ×2subscribed ×2commented ×1

Fix Action

Fixed

Fixed by PR: fix: session grouping in Spend Logs UI for query_raw dict rows (https://github.com/BerriAI/litellm/pull/23245)

PR fix notes

PR #23245: fix: session grouping in Spend Logs UI for query_raw dict rows

Repository: BerriAI/litellm
Author: naaa760
State: open | merged: False
Link: https://github.com/BerriAI/litellm/pull/23245

Description (problem / solution / changelog)

fix : #23157

relevant issues

Fixes Spend Logs session grouping.

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR [ ] I have Added testing in the directory, Adding at least 1 test is a hard requirement - see details [ ] My PR passes all unit tests on [ ] My PR's scope is as isolated as possible, it only solves 1 specific problem [ ] I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

CI (LiteLLM team)

CI status guideline:

50-55 passing tests: main is stable with minor issues.

45-49 passing tests: acceptable but needs attention

<= 40 passing tests: unstable; be careful with your merges and assess the risk.

Type

🐛 Bug Fix

Changes

Fix session grouping.
Fix session_total_count.
Add test.

Changed files

litellm/proxy/spend_tracking/spend_management_endpoints.py (modified, +27/-12)

Code Example

export LITELLM_URL=http://localhost:4000
export LITELLM_KEY=someKey
export MODEL="openrouter/google/gemini-3-flash-preview" # use any configured model
export SESSION_ID="repro-session-123"

for i in 1 2 3; do
  curl -sS "$LITELLM_URL/chat/completions" \
    -H "Authorization: Bearer $LITELLM_KEY" \
    -H "Content-Type: application/json" \
    -d "{
      \"model\": \"$MODEL\",
      \"messages\": [{\"role\": \"user\", \"content\": \"repro $i\"}],
      \"litellm_session_id\": \"$SESSION_ID\"
    }" > /dev/null
done

---

curl -sS \
  -H "Authorization: Bearer $LITELLM_KEY" \
  "$LITELLM_URL/spend/logs/session/ui?session_id=$SESSION_ID" | jq

---

curl -sS \
  -H "Authorization: Bearer $LITELLM_KEY" \
  "$LITELLM_URL/spend/logs/ui?start_date=2026-03-09%2000:00:00&end_date=2026-03-10%2000:00:00&page=1&page_size=50" \
  | jq '.data | map(select(.session_id == "repro-session-123")) | map({request_id, session_id, session_total_count})'

RAW_BUFFERClick to expand / collapse

Check for existing issues

I have searched the existing issues and checked that my issue is not a duplicate.

What happened?

The above all have the exact same sessionId - chat-preview:c896735d087144350aa43c8b2da0deca9 but whenever you click on the session:

They are not grouped at all.

Looks like this was introduced in https://github.com/BerriAI/litellm/pull/21018 by @ishaan-jaff ( incredible work on this product btw, if you read this!! )

Steps to Reproduce (Codex put this together, but I did verify manually that it repro'd)

Repro Steps

Environment:

LiteLLM version with bug: v1.81.12-stable.1 Working comparison: 1.80.15-stable.1 Observed on POST /chat/completions and the Logs UI / /spend/logs/ui Start LiteLLM v1.81.12-stable.1 with spend logging enabled and a DB attached.

Send 3 chat completion requests with the same litellm_session_id.

export LITELLM_URL=http://localhost:4000
export LITELLM_KEY=someKey
export MODEL="openrouter/google/gemini-3-flash-preview" # use any configured model
export SESSION_ID="repro-session-123"

for i in 1 2 3; do
  curl -sS "$LITELLM_URL/chat/completions" \
    -H "Authorization: Bearer $LITELLM_KEY" \
    -H "Content-Type: application/json" \
    -d "{
      \"model\": \"$MODEL\",
      \"messages\": [{\"role\": \"user\", \"content\": \"repro $i\"}],
      \"litellm_session_id\": \"$SESSION_ID\"
    }" > /dev/null
done

Confirm the dedicated session endpoint returns all 3 rows.

curl -sS \
  -H "Authorization: Bearer $LITELLM_KEY" \
  "$LITELLM_URL/spend/logs/session/ui?session_id=$SESSION_ID" | jq

Expected result: total = 3 data contains all 3 requests

Now fetch the normal logs list endpoint for a time range that includes those requests.

curl -sS \
  -H "Authorization: Bearer $LITELLM_KEY" \
  "$LITELLM_URL/spend/logs/ui?start_date=2026-03-09%2000:00:00&end_date=2026-03-10%2000:00:00&page=1&page_size=50" \
  | jq '.data | map(select(.session_id == "repro-session-123")) | map({request_id, session_id, session_total_count})'

Actual result on v1.81.12-stable.1:

all matching rows have session_total_count: 1 Expected result:

Each matching row should have session_total_count: 3
the Logs UI should treat them as one grouped session

Open the Logs UI and click the session column for one of those rows. Actual: rows are not grouped as a multi-request session Expected: they should open as one grouped session

What part of LiteLLM is this about?

UI Dashboard

What LiteLLM version are you on ?

v1.80.15-stable.1

extent analysis

Fix Plan

To resolve the issue of session grouping in the Logs UI, we need to update the logic for calculating session_total_count and grouping sessions.

Here are the steps:

Update the database query to correctly calculate session_total_count for each session.
Modify the Logs UI to group sessions based on the session_id.
Ensure that the session_id is correctly passed and used in the API requests.

Code Changes

# Update the database query to calculate session_total_count
def get_session_total_count(session_id):
    # Assuming a database connection is established
    query = "SELECT COUNT(*) FROM logs WHERE session_id = %s"
    cursor.execute(query, (session_id,))
    return cursor.fetchone()[0]

# Modify the Logs UI to group sessions
def get_grouped_sessions(logs):
    grouped_sessions = {}
    for log in logs:
        session_id = log['session_id']
        if session_id not in grouped_sessions:
            grouped_sessions[session_id] = []
        grouped_sessions[session_id].append(log)
    return grouped_sessions

# Ensure correct session_id usage in API requests
def get_logs(session_id):
    # Use the session_id to fetch logs
    query = "SELECT * FROM logs WHERE session_id = %s"
    cursor.execute(query, (session_id,))
    return cursor.fetchall()

Verification

To verify the fix, follow these steps:

Send multiple chat completion requests with the same litellm_session_id.
Check the dedicated session endpoint to ensure all requests are returned.
Fetch the normal logs list endpoint for a time range that includes those requests.
Verify that each matching row has the correct session_total_count.
Open the Logs UI and click the session column for one of those rows to ensure they are grouped correctly.

Extra Tips

Ensure that the database connection is stable and queries are optimized for performance.
Consider adding logging and error handling to the updated code to improve debugging and user experience.
Test the fix thoroughly to ensure it resolves the issue without introducing new problems.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #ssr #installation #tensor shape #autograd error #indexing error #inference speed #output truncation #response parsing #generation error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

litellm - ✅(Solved) Fix [Bug]: Sessions are no longer correctly grouped together in the Spend Logs UI [1 pull requests, 1 comments, 2 participants]

Recommended Tools

GitHub issue graph ai analysis

Fix Action

Fixed

PR fix notes

PR #23245: fix: session grouping in Spend Logs UI for query_raw dict rows

Description (problem / solution / changelog)

relevant issues

Pre-Submission checklist

CI (LiteLLM team)

Type

Changes

Changed files

Code Example

Check for existing issues

What happened?

Steps to Reproduce (Codex put this together, but I did verify manually that it repro'd)

What part of LiteLLM is this about?

What LiteLLM version are you on ?

extent analysis

Fix Plan

Code Changes

Verification

Extra Tips

Still need to ship something?

TRENDING

litellm - ✅(Solved) Fix [Bug]: Sessions are no longer correctly grouped together in the Spend Logs UI [1 pull requests, 1 comments, 2 participants]

Recommended Tools

GitHub issue graph ai analysis

Fix Action

Fixed

PR fix notes

PR #23245: fix: session grouping in Spend Logs UI for query_raw dict rows

Description (problem / solution / changelog)

relevant issues

Pre-Submission checklist

CI (LiteLLM team)

Type

Changes

Changed files

Code Example

Check for existing issues

What happened?

Steps to Reproduce (Codex put this together, but I did verify manually that it repro'd)

What part of LiteLLM is this about?

What LiteLLM version are you on ?

extent analysis

Fix Plan

Code Changes

Verification

Extra Tips

Still need to ship something?

RELATED_DISCOVERY

TRENDING